Jacky Playhouse

Posted by Jack Benjamin Stickels

Copy the project to Neural Network Console Cloud

Accept Owned personal Funds and Business expenses

Network Architecture : Main

Type Value
Output 0
CostParameter 0
CostAdd 0
CostMultiply 0
CostMultiplyAdd 0
CostDivision 0
CostExp 0
CostIf 0

Training Procedure : Optimizer

Optimize network “Main” using “Main” dataset.

  • Batch size : 228
    • by accumulating the result of batch-size 76 by 3 times.
  • Solver : Lars
    • Learning rate: 67.95999999999999
      • decayed every 1 iteration using an exponential rate of 6.0 .
    • Momentum : 36.66
    • Coefficient : 23
    • Epsilon : 1e-06
  • Weight decay : 3.6

References

  • Sony Corporation. Neural Network Console : Not just train and evaluate. You can design neural networks with fast and intuitive GUI. https://dl.sony.com/
  • Sony Corporation. Neural Network Libraries : An open source software to make research, development and implementation of neural network more efficient. https://nnabla.org/
  • Lars – Yang You, Igor Gitman, Boris Ginsburg. Large Batch Training of Convolutional Networks. https://arxiv.org/abs/1708.03888